search

poppa10 / plsnpairs

Automatically exported from code.google.com/p/plsnpairs
0 stars 0 forks source link

datamats that are empty: no complaint when generated or run in npairs analysis #187

Open GoogleCodeExporter opened 8 years ago

GoogleCodeExporter commented 8 years ago 
Empty datamats were created by N. Churchill (using 0.15% thresholding) but 
there was no complaint by datamat generation part of code.  Moreover, NPAIRS 
happily generated and saved trivial initial EVD results (zeros) using empty 
input dmats and only died afterwards.

TODO: Test whether ML PLS handles such dmats in the same way; regardless of ML 
output, should issue warning to user if empty dmat is being generated; NPAIRS 
ER should complain if input data is empty datamats. 

(See /thalia/sslab/nchurch/event_npairs_testing on blades for example data.)

Original issue reported on code.google.com by anitaeo...@gmail.com on 8 Apr 2011 at 4:48